College Preview

When It Comes To Defense, Efficiency Matters




Follow me on Twitter


See also: New Bats Make Defense Even More Important


At the end of every season, Creighton coach Ed Servais crunches the numbers.

"Because numbers are stories," he said. "We try to incorporate what we learn from those numbers into our practice."

Through studying a sample of more than 400 games, Servais has ascertained that the Bluejays win 88 percent of their games when they hold opponents to three runs or fewer. When they hold opponents to five or fewer runs, they win 78 percent of the time.

2010 Defensive Efficiency
1. South Carolina .726
2. Texas .726
3. Virginia .717
4. UCLA .713
5. Rice .711
6. Louisiana-Lafayette .710
7. Stanford .704
8. Toledo .704
9. Miami .701
10. Connecticut .701
11. Nevada .700
12. Jackson State .699
13. Coastal Carolina .699
14. St. John's .698
15. Oregon State .698
16. Texas Christian .697
17. Cal State Fullerton .696
18. Bryant .695
19. Portland .695
20. Le Moyne .694
21. Arizona State .693
22. Tennessee .692
23. Indiana State .692
24. Florida State .691
25. Washington State .690
26. Creighton .690
27. Charlotte .690
28. Wake Forest .690
29. Virginia Commonwealth .690
30. Louisville .690
"I think that's a large enough sample that you can sell your players on things that actually happen," Servais said. "Once we start throwing those numbers at them, they start to buy in. They're interested in winning."

Who isn't? Statistics are wonderful tools to help fans understand the game better, but they can also be used to help teams win more—even simply by demonstrating to players the importance of run prevention and good defense.

Defensive statistics are often regarded as the holy grail of baseball stats. Even at the major league level, defense is not easy to quantify, but it's particularly difficult at the college level, where television broadcasts and cutting-edge technology is rare, and the sample size is smaller. That doesn't stop Duke coach Sean McNally from calculating his players' range factors on his own.

"That's an easy one—I can measure based on guys we've had in the past," he said. "We do a lot of it ourselves. I just love studying stats, comparing the shortstops in our league, for instance. We'll compare guys from year to year: Did (former Duke shortstop) Jake Lemmerman get to more balls as a junior than as a sophomore, and if so, why? You can get lost in all the data, but we try to find as much as we can, anything that's out there. Defense is the toughest to measure, but you've got to measure what you can."

The NCAA publishes leaders for two defensive stat categories: fielding percentage and double plays. Fielding percentage can give you an idea how good teams are at making routine plays, but it doesn't show which teams are actually making the most plays, according to Jeff Sackmann, who founded collegesplits.com along with Kent Bonham in 2006.

"The standard critique of traditional fielding stats is they don't give you the whole picture, because you could make a lot of errors and still be a good fielder," Sackmann said. "Like Mike Bordick might not make as many errors as Ozzie Smith, but that doesn't mean he's better. The idea is to find out how many plays they're making."

2010 Fielding Percentage
1. Texas .980
2. Creighton .980
3. Boston College .979
4. San Francisco .978
5. Oklahoma State .978
6. Florida .978
7. Virginia .977
8. Alabama .977
9. Pittsburgh .977
10. Indiana State .976
11. Oklahoma .976
12. Arizona State .976
13. South Carolina .975
14. Troy .975
15. Evansville .975
16. Villanova .975
17. Duke .975
18. Louisiana State .975
19. Michigan State .975
20. Vanderbilt .975
21. Dartmouth .975
22. Mississippi .974
23. UC Irvine .974
24. Tennessee .974
25. Rice .974
26. Rider .974
27. Miami (Ohio) .974
28. Oregon State .973
29. Kent State .973
30. Minnesota .973
To do that, College Splits calculates defensive efficiency for every Division I team. Sackmann spent the better part of five years perfecting a computer code that translates the play-by-play descriptions that accompany college box scores into usable data. That gives him a record of every play that is made over the course of the Division I season, which he can use to compute each team's defensive efficiency, among plenty of other stats.

"Defensive efficiency will give you an idea of the team's defense as a whole," he said. "So a defensive efficiency rating of .700 means that any time a ball comes off the bat and doesn't go for a home run, 70 percent of the time a team's turning it into an out. So .700 is pretty good; bad teams can get down to about .660, and sometimes in college baseball you'll see it get even lower. There are some pretty ham-handed teams, especially in small colleges."

Of course, Sackmann said the numbers should be taken "with a giant grain of salt," because they don't tell the entire story, either.

"We know that not every ball coming off the bat is equally easy to field, so that's why coaches and pitchers love ground balls—those are a lot easier to field than anything else," Sackmann said. "If a staff is churning out ground ball after ground ball, you might end up with a good defensive efficiency because the pitchers are making it easier for them. So it's tricky to tease out how much of it is luck, how much of it is the pitching staff, and how much of it is a good third baseman or shortstop."

Viewed together, defensive efficiency and fielding percentage can paint a pretty good picture about which teams are the best at defending.

College Splits pays its bills by providing information to its clients—largely major league clubs. At the end of last season, Sackmann said he and Bonham had contracts with 16 different clubs, and they expect to work with 15-20 this year.

Major league scouting departments want to know about college players, not college teams. When it comes to defense, College Splits aims to determine how many plays a particular shortstop makes relative to an average shortstop, for instance.

"We look at how many ground balls a specific team is getting, then how many plays a player at that position is making," Sackmann said. "We do that for all of college baseball—how many plays a left fielder is making per fly ball, etc. Those numbers in themselves aren't very meaningful. If I told you that a player made plays on 12 percent of ground balls by his pitching staff, that wouldn't mean much. But if we said the average NCAA shortstop made plays on 16 percent of ground balls, we'd know that player isn't making that many plays. It shows he's either got a slow first step, or he's got a weak arm, or he isn't beating guys to the bag—it could be anything."

Sackmann doesn't suggest that his statistics can function as alternatives to scouting. But he says his clients use the stats as feedback against what scouts are saying, and they use the scouts as feedback against what the stats are saying.

The more tools available to measure defense, the better.

Below is a list of the top defenders in Division I last year at three positions: third base, shortstop and center field. Sackmann said there is more "noise" in the data for other positions, where defenders might not get as many defensive chances. Players are sorted by the number of plays they made more than an average defender at their position made. Asterisks denote players who are still on college rosters heading into 2011.

Shortstop
Player School Plays Above Average
*A.J. Rusbarsky Seton Hall 18
*Austin Nola Louisiana State 15
*Evan Boyd Mercer 12
*Stephen Wickens Florida Gulf Coast 11
*Eric Stamets Evansville 11
*Neal Pritchard Elon 11
Blake Kelso Houston 10
*Aaron Miller Gardner-Webb 10
*Kenton Parmley Southeast Mo. State 9
Kevin Haas Akron 9
 
Third base
Player School Plays Above Average
*Jason Esposito Vanderbilt 10
Tyler Burnett Middle Tennessee State 9
Carlos Alonso Delaware 8
*Effrey Valdez New York Tech 7
*Mark Threlkeld Louisiana Tech 7
Mike Olt Connecticut 7
*Matt Duffy Tennessee 7
Mike Walker Pacific 7
*Jake Overstreet South Alabama 6
Nick Kuroczko Utah 6
 
Center field
Player School Plays Above Average
Westley Moss Nevada 16
J.D. Dunn St. Louis 12
*Jason Martin San Jose State 12
*Elliot Frey Murray State 12
Robbie Anston Boston College 11
*Taylor Dugas Alabama 10
Brian Heere Kansas 9
Justin Bencsko Villanova 9
*Chad Hinshaw Illinois State 9
Ryan Strausborger Indiana State 8
* Returns for 2011